app.py

#!/usr/bin/env python3
"""
Production-Ready Gradio Chat Interface

This is the main application file for production deployment of the
learning conversational travel chat system.

Features:
- Environment-specific configuration
- Comprehensive error handling
- Rate limiting and session management
- Security measures and input validation
- Production-ready logging and monitoring

Author: AI Assistant
Date: 2024
"""

import os
import sys
import logging
import asyncio
from typing import Dict, List, Any, Optional, Tuple
import gradio as gr
from datetime import datetime
import json

# Ensure the app directory itself is importable (for HF runtime quirks)
APP_DIR = os.path.dirname(__file__)
if APP_DIR not in sys.path:
    sys.path.insert(0, APP_DIR)

# Add the src directory to the path
sys.path.insert(0, os.path.join(os.path.dirname(__file__), 'src'))

# Import configuration and security systems
from config.settings import get_config, is_production, is_development

# Initialize a basic logger early so it's available during import fallbacks
logger = logging.getLogger(__name__)

# Try to import security systems, with fallback for Hugging Face deployment
try:
    from security.error_handler import (
        InputValidator, RateLimiter, SessionManager, ErrorHandler,
        handle_errors, validate_input, rate_limit, require_session,
        ValidationError, RateLimitError, AuthenticationError
    )
    SECURITY_AVAILABLE = True
except ImportError as e:
    logger.warning(f"Security modules not available: {e}")
    logger.warning("Running in Hugging Face compatibility mode")
    SECURITY_AVAILABLE = False
    
    # Create fallback classes for Hugging Face deployment
    class InputValidator:
        def __init__(self, max_length=1000):
            self.max_length = max_length

        def _truncate(self, text: str) -> str:
            return (text or "")[: self.max_length]

        def validate(self, text: str) -> str:
            return self._truncate(text)

        def validate_user_id(self, user_id: str) -> str:
            return self._truncate(user_id)

        def validate_message(self, message: str) -> str:
            return self._truncate(message)

        def validate_input(self, value: str, _name: str = "") -> str:
            return self._truncate(value)
    
    class RateLimiter:
        def __init__(self, requests_per_minute=60, requests_per_hour=1000):
            self.requests_per_minute = requests_per_minute
            self.requests_per_hour = requests_per_hour

        def check_rate_limit(self, user_id: str) -> bool:
            return True

        def get_remaining_requests(self, _user_id: str) -> dict:
            return {
                "per_minute_remaining": self.requests_per_minute,
                "per_hour_remaining": self.requests_per_hour,
            }
    
    class SessionManager:
        def __init__(self, session_timeout_minutes=30):
            self.session_timeout_minutes = session_timeout_minutes
            self.sessions: Dict[str, dict] = {}

        def create_session(self, user_id: str, data: Optional[dict] = None) -> str:
            session_id = f"session_{user_id}_{datetime.now().timestamp()}"
            self.sessions[session_id] = {
                "user_id": user_id,
                "created_at": datetime.now(),
                "last_activity": datetime.now(),
                "data": data or {},
            }
            return session_id

        def validate_session(self, session_id: str) -> Optional[dict]:
            return self.sessions.get(session_id)

        def destroy_session(self, session_id: str) -> None:
            self.sessions.pop(session_id, None)
    
    class ErrorHandler:
        def __init__(self):
            self._errors: List[dict] = []

        def handle_error(self, error, context: str = ""):
            logger.error(f"Error: {error} - Context: {context}")
            self._errors.append({
                "error": str(error),
                "context": context,
                "time": datetime.now().isoformat(),
            })
            return "An error occurred. Please try again."

        def get_error_statistics(self) -> dict:
            return {"total_errors": len(self._errors)}
    
    # Fallback decorators
    def handle_errors(func):
        def wrapper(*args, **kwargs):
            try:
                return func(*args, **kwargs)
            except Exception as e:
                logger.error(f"Error in {func.__name__}: {e}")
                return "An error occurred. Please try again."
        return wrapper
    
    def validate_input(func):
        return func
    
    def rate_limit(func):
        return func
    
    def require_session(func):
        return func
    
    # Fallback exceptions
    class ValidationError(Exception):
        pass
    class RateLimitError(Exception):
        pass
    class AuthenticationError(Exception):
        pass

# Import the real API chat system for Hugging Face deployment (robust fallback)
try:
    from conversational_travel_chat_real_apis import RealAPIConversationalTravelChat
except ModuleNotFoundError as e:
    logger.warning(f"Primary import failed: {e}. sys.path={sys.path}")
    # Attempt dynamic import via importlib as a fallback
    try:
        import importlib.util
        module_path = os.path.join(APP_DIR, 'conversational_travel_chat_real_apis.py')
        spec = importlib.util.spec_from_file_location('conversational_travel_chat_real_apis', module_path)
        module = importlib.util.module_from_spec(spec)
        assert spec and spec.loader
        spec.loader.exec_module(module)
        RealAPIConversationalTravelChat = getattr(module, 'RealAPIConversationalTravelChat')
        logger.info("Loaded RealAPIConversationalTravelChat via dynamic import")
    except Exception as inner_e:
        logger.error(f"Failed to load conversational_travel_chat_real_apis: {inner_e}")
        raise

# Set up logging
logging.basicConfig(
    level=logging.INFO,
    format='%(asctime)s - %(name)s - %(levelname)s - %(message)s',
    handlers=[
        logging.StreamHandler(),
        logging.FileHandler('app.log')
    ]
)

logger = logging.getLogger(__name__)

class ProductionChatApp:
    """
    Production-ready chat application with security and monitoring
    """
    
    def __init__(self):
        # Load configuration
        self.config = get_config()
        
        # Initialize security systems (with fallback for Hugging Face)
        if SECURITY_AVAILABLE:
            self.input_validator = InputValidator(max_length=self.config.security.max_input_length)
            self.rate_limiter = RateLimiter(
                requests_per_minute=self.config.api.rate_limit_per_minute,
                requests_per_hour=self.config.api.rate_limit_per_hour
            )
            self.session_manager = SessionManager(
                session_timeout_minutes=self.config.security.session_timeout_minutes
            )
            self.error_handler = ErrorHandler()
        else:
            # Use fallback security systems for Hugging Face
            self.input_validator = InputValidator(max_length=1000)
            self.rate_limiter = RateLimiter(requests_per_minute=60, requests_per_hour=1000)
            self.session_manager = SessionManager(session_timeout_minutes=30)
            self.error_handler = ErrorHandler()
        
        # Initialize real API chat system for Hugging Face
        self.real_api_chat = RealAPIConversationalTravelChat()
        
        # Application state
        self.active_sessions = {}
        self.request_count = 0
        
        # Check for required API keys for Hugging Face deployment
        self._check_api_keys()
        
        logger.info(f"Production chat app initialized in {self.config.deployment.environment.value} mode")
    
    def _check_api_keys(self):
        """Check for required API keys for Hugging Face deployment"""
        required_vars = ['SERPAPI_API_KEY', 'TAVILY_API_KEY']
        missing_vars = [var for var in required_vars if not os.getenv(var)]
        
        if missing_vars:
            logger.warning(f"Missing environment variables: {missing_vars}")
            logger.warning("The app will run with limited functionality.")
        else:
            logger.info("All required API keys are configured")
    
    @handle_errors
    async def start_conversation(self, user_id: str, message: str, session_id: Optional[str] = None) -> Tuple[str, str, str]:
        """Start or continue a conversation with security checks"""
        
        # Validate inputs
        user_id = self.input_validator.validate_user_id(user_id)
        message = self.input_validator.validate_message(message)
        
        # Check rate limiting
        if not self.rate_limiter.check_rate_limit(user_id):
            raise RateLimitError("Rate limit exceeded. Please wait before making another request.")
        
        # Handle session management
        if session_id:
            session_data = self.session_manager.validate_session(session_id)
            if not session_data:
                raise AuthenticationError("Invalid or expired session")
        else:
            session_id = self.session_manager.create_session(user_id, {
                "start_time": datetime.now().isoformat(),
                "message_count": 0
            })
        
        # Update session data
        session_data = self.session_manager.sessions[session_id]
        session_data["data"]["message_count"] = session_data["data"].get("message_count", 0) + 1
        session_data["last_activity"] = datetime.now()
        
        # Process conversation with real API chat
        try:
            response = await self.real_api_chat.process_message(user_id, message)
            
            # Update request count
            self.request_count += 1
            
            # Log successful interaction
            logger.info(f"Conversation processed for user {user_id}, session {session_id}")
            
            return response, session_id, "success"
            
        except Exception as e:
            logger.error(f"Error processing conversation for user {user_id}: {e}")
            raise
    
    @handle_errors
    def collect_feedback(self, user_id: str, session_id: str, rating: int, comments: str = "") -> str:
        """Collect user feedback with validation"""
        
        # Validate inputs
        user_id = self.input_validator.validate_user_id(user_id)
        session_id = self.input_validator.validate_input(session_id, "session_id")
        
        # Validate rating
        if not isinstance(rating, int) or rating < 1 or rating > 5:
            raise ValidationError("Rating must be an integer between 1 and 5")
        
        # Validate comments if provided
        if comments:
            comments = self.input_validator.validate_input(comments, "comments")
        
        # Validate session
        session_data = self.session_manager.validate_session(session_id)
        if not session_data:
            raise AuthenticationError("Invalid or expired session")
        
        # Collect feedback
        self.learning_chat.collect_feedback(user_id, rating, comments)
        
        logger.info(f"Feedback collected from user {user_id}: {rating}/5")
        return f"Thank you for your feedback! Rating: {rating}/5"
    
    @handle_errors
    def end_conversation(self, user_id: str, session_id: str, completion_status: str = "completed") -> str:
        """End user conversation"""
        
        # Validate inputs
        user_id = self.input_validator.validate_user_id(user_id)
        session_id = self.input_validator.validate_input(session_id, "session_id")
        
        # Validate completion status
        valid_statuses = ["completed", "abandoned", "in_progress"]
        if completion_status not in valid_statuses:
            raise ValidationError(f"Completion status must be one of: {valid_statuses}")
        
        # Validate session
        session_data = self.session_manager.validate_session(session_id)
        if not session_data:
            raise AuthenticationError("Invalid or expired session")
        
        # End conversation
        self.learning_chat.end_conversation(user_id, completion_status)
        
        # Destroy session
        self.session_manager.destroy_session(session_id)
        
        logger.info(f"Conversation ended for user {user_id} with status: {completion_status}")
        return f"Session ended successfully. Status: {completion_status}"
    
    @handle_errors
    def get_analytics(self, admin_key: str) -> str:
        """Get system analytics (admin only)"""
        
        # Simple admin authentication (in production, use proper auth)
        if admin_key != os.getenv('ADMIN_KEY', 'admin123'):
            raise AuthenticationError("Invalid admin key")
        
        # Get analytics
        insights = self.learning_chat.get_learning_insights()
        
        # Format for display
        analytics_data = {
            "timestamp": datetime.now().isoformat(),
            "request_count": self.request_count,
            "active_sessions": len(self.session_manager.sessions),
            "rate_limit_stats": self.rate_limiter.get_remaining_requests("system"),
            "error_stats": self.error_handler.get_error_statistics(),
            "learning_insights": insights
        }
        
        return json.dumps(analytics_data, indent=2, default=str)
    
    @handle_errors
    def health_check(self) -> Dict[str, Any]:
        """Health check endpoint"""
        return {
            "status": "healthy",
            "timestamp": datetime.now().isoformat(),
            "environment": self.config.deployment.environment.value,
            "version": "1.0.0",
            "uptime": "active"
        }
    
    def create_interface(self) -> gr.Blocks:
        """Create the Gradio interface"""
        
        # Custom CSS for production styling
        custom_css = """
        .gradio-container {
            max-width: 1200px !important;
            margin: 0 auto !important;
        }
        .chat-message {
            padding: 10px;
            margin: 5px 0;
            border-radius: 10px;
        }
        .user-message {
            background-color: #e3f2fd;
            text-align: right;
        }
        .bot-message {
            background-color: #f5f5f5;
            text-align: left;
        }
        .error-message {
            background-color: #ffebee;
            color: #c62828;
            border: 1px solid #ef5350;
        }
        .success-message {
            background-color: #e8f5e8;
            color: #2e7d32;
            border: 1px solid #4caf50;
        }
        #send-button {
            background: linear-gradient(45deg, #4CAF50, #45a049) !important;
            border: none !important;
            color: white !important;
            font-weight: bold !important;
            transition: all 0.3s ease !important;
        }
        #send-button:hover {
            background: linear-gradient(45deg, #45a049, #4CAF50) !important;
            transform: translateY(-2px) !important;
            box-shadow: 0 4px 8px rgba(0,0,0,0.2) !important;
        }
        .textbox textarea {
            border: 2px solid #e0e0e0 !important;
            border-radius: 8px !important;
            transition: border-color 0.3s ease !important;
        }
        .textbox textarea:focus {
            border-color: #4CAF50 !important;
            outline: none !important;
            box-shadow: 0 0 0 3px rgba(76, 175, 80, 0.1) !important;
        }
        .textbox textarea {
            resize: vertical !important;
        }
        """
        
        with gr.Blocks(
            title="AI Travel Assistant",
            css=custom_css,
            theme=gr.themes.Soft() if is_development() else gr.themes.Default()
        ) as interface:
            
            gr.Markdown("""
            # 🧠 AI Travel Assistant
            
            Your intelligent travel planning companion that learns from every interaction
            to provide increasingly personalized experiences.
            
            **Features:**
            - 🎯 Personalized recommendations based on your travel style
            - 📊 Continuous learning from your preferences
            - 🔒 Secure and private conversations
            - 📈 Real-time feedback and improvement
            """)
            
            with gr.Tab("💬 Chat"):
                with gr.Row():
                    with gr.Column(scale=1):
                        # User input section
                        user_id_input = gr.Textbox(
                            label="User ID",
                            placeholder="Enter your user ID (e.g., user_123)",
                            value="user_123",
                            interactive=True
                        )
                        
                        session_id_input = gr.Textbox(
                            label="Session ID",
                            placeholder="Session ID (auto-generated)",
                            interactive=False,
                            visible=False
                        )
                        
                        gr.Markdown("### Start Your Conversation")
                        
                        chatbot = gr.Chatbot(
                            label="Conversation",
                            height=400,
                            show_label=False,
                            container=True
                        )
                        
                        message_input = gr.Textbox(
                            label="Message",
                            placeholder="Tell me about your travel plans... (Press Enter to send)",
                            lines=2,
                            max_lines=4,
                            show_copy_button=False,
                            container=True,
                            scale=4
                        )
                        
                        with gr.Row():
                            send_btn = gr.Button("Send ⏎", variant="primary", scale=1, elem_id="send-button")
                            clear_btn = gr.Button("Clear", variant="secondary", scale=1)
                        
                        gr.Markdown("💡 **Tip:** Press Enter to send your message quickly!")
                    
                    with gr.Column(scale=1):
                        # Feedback and session management
                        gr.Markdown("### 📊 Feedback & Session")
                        
                        with gr.Row():
                            feedback_rating = gr.Slider(
                                minimum=1, maximum=5, step=1, value=3,
                                label="Rate your experience (1-5)"
                            )
                        
                        feedback_comments = gr.Textbox(
                            label="Comments (optional)",
                            placeholder="Tell us how we can improve...",
                            lines=3,
                            max_lines=5
                        )
                        
                        with gr.Row():
                            feedback_btn = gr.Button("Submit Feedback", variant="secondary")
                        
                        gr.Markdown("### 🔧 Session Management")
                        
                        completion_status = gr.Dropdown(
                            choices=["completed", "abandoned", "in_progress"],
                            value="completed",
                            label="Session Status"
                        )
                        
                        end_session_btn = gr.Button("End Session", variant="stop")
                        
                        # Status display
                        status_display = gr.Textbox(
                            label="Status",
                            interactive=False,
                            lines=2
                        )
            
            with gr.Tab("📊 Analytics"):
                gr.Markdown("### 📈 System Analytics")
                
                admin_key_input = gr.Textbox(
                    label="Admin Key",
                    placeholder="Enter admin key to view analytics",
                    type="password"
                )
                
                analytics_btn = gr.Button("View Analytics", variant="primary")
                analytics_output = gr.JSON(label="Analytics Data")
            
            with gr.Tab("🔧 Health Check"):
                gr.Markdown("### 🏥 System Health")
                
                health_btn = gr.Button("Check Health", variant="primary")
                health_output = gr.JSON(label="Health Status")
            
            # Event handlers
            async def process_message(user_id, message, session_id, history):
                """Process user message with error handling"""
                try:
                    if not message.strip():
                        return "", history, session_id, "Please enter a message."
                    
                    response, new_session_id, status = await self.start_conversation(user_id, message, session_id)
                    
                    # Update history
                    history.append([message, response])
                    
                    return "", history, new_session_id, f"✅ {status}"
                    
                except Exception as e:
                    error_msg = f"❌ Error: {str(e)}"
                    return "", history, session_id, error_msg
            
            def clear_conversation():
                """Clear conversation history"""
                return [], None, "Conversation cleared."
            
            def submit_feedback(user_id, session_id, rating, comments):
                """Submit user feedback"""
                try:
                    if not session_id:
                        return "❌ No active session. Please start a conversation first."
                    
                    result = self.collect_feedback(user_id, session_id, rating, comments)
                    return f"✅ {result}"
                except Exception as e:
                    return f"❌ Error: {str(e)}"
            
            def end_session(user_id, session_id, status):
                """End user session"""
                try:
                    if not session_id:
                        return "❌ No active session to end."
                    
                    result = self.end_conversation(user_id, session_id, status)
                    return f"✅ {result}"
                except Exception as e:
                    return f"❌ Error: {str(e)}"
            
            def view_analytics(admin_key):
                """View system analytics"""
                try:
                    result = self.get_analytics(admin_key)
                    return result
                except Exception as e:
                    return f"❌ Error: {str(e)}"
            
            def check_health():
                """Check system health"""
                try:
                    result = self.health_check()
                    return result
                except Exception as e:
                    return {"status": "error", "message": str(e)}
            
            # Connect events - Enter key and Send button both trigger the same function
            def handle_message_submit(user_id, message, session_id, history):
                """Handle message submission from either Enter key or Send button"""
                print(f"🔍 DEBUG: handle_message_submit called with message: '{message}'")
                if not message.strip():
                    print("🔍 DEBUG: Empty message detected")
                    return "", history, session_id, "Please enter a message."
                
                print(f"🔍 DEBUG: Processing message: '{message}'")
                # Call the async process_message function and return its result
                import asyncio
                try:
                    loop = asyncio.get_event_loop()
                    result = loop.run_until_complete(process_message(user_id, message, session_id, history))
                    return result
                except RuntimeError:
                    # If no event loop is running, create a new one
                    result = asyncio.run(process_message(user_id, message, session_id, history))
                    return result
            
            # Send button click
            send_btn.click(
                handle_message_submit,
                inputs=[user_id_input, message_input, session_id_input, chatbot],
                outputs=[message_input, chatbot, session_id_input, status_display]
            )
            
            # Enter key press in message input - Enhanced configuration
            message_input.submit(
                handle_message_submit,
                inputs=[user_id_input, message_input, session_id_input, chatbot],
                outputs=[message_input, chatbot, session_id_input, status_display],
                show_progress=False,
                queue=True
            )
            
            clear_btn.click(
                clear_conversation,
                outputs=[chatbot, session_id_input, status_display]
            )
            
            feedback_btn.click(
                submit_feedback,
                inputs=[user_id_input, session_id_input, feedback_rating, feedback_comments],
                outputs=[status_display]
            )
            
            end_session_btn.click(
                end_session,
                inputs=[user_id_input, session_id_input, completion_status],
                outputs=[status_display]
            )
            
            analytics_btn.click(
                view_analytics,
                inputs=[admin_key_input],
                outputs=[analytics_output]
            )
            
            health_btn.click(
                check_health,
                outputs=[health_output]
            )
        
        return interface
    
    def run(self):
        """Run the application"""
        interface = self.create_interface()
        
        # Detect Hugging Face Spaces environment
        running_in_hf_space = bool(os.getenv("SPACE_ID") or os.getenv("HF_SPACE_ID") or os.getenv("HF_HOME"))

        # Configure host/port for Spaces explicitly
        host = "0.0.0.0" if running_in_hf_space else self.config.deployment.host
        port = int(os.getenv("PORT", "7860")) if running_in_hf_space else self.config.deployment.port

        # Launch (avoid SSL params in Spaces; enable queue)
        launch_kwargs = dict(
            server_name=host,
            server_port=port,
            share=False,
            debug=self.config.deployment.debug_mode,
            show_error=True,
            quiet=not self.config.deployment.debug_mode,
        )

        # Only pass SSL options outside Spaces
        if not running_in_hf_space:
            launch_kwargs.update(
                ssl_certfile=self.config.deployment.ssl_cert_path,
                ssl_keyfile=self.config.deployment.ssl_key_path,
                ssl_verify=False if is_development() else True,
            )

        interface.queue().launch(**launch_kwargs)

def main():
    """Main application entry point"""
    try:
        # Initialize and run the application
        app = ProductionChatApp()
        
        logger.info("Starting production chat application...")
        logger.info(f"Environment: {app.config.deployment.environment.value}")
        logger.info(f"Host: {app.config.deployment.host}")
        logger.info(f"Port: {app.config.deployment.port}")
        logger.info(f"Debug mode: {app.config.deployment.debug_mode}")
        
        app.run()
        
    except Exception as e:
        logger.error(f"Failed to start application: {e}")
        sys.exit(1)

if __name__ == "__main__":
    main()

# Expose a global Gradio interface for Hugging Face Spaces autodiscovery
try:
    if os.getenv("SPACE_ID") or os.getenv("HF_SPACE_ID") or os.getenv("HF_HOME"):
        _app_instance = ProductionChatApp()
        demo = _app_instance.create_interface()
except Exception:
    # Do not fail module import on Spaces if lazy creation is desired
    pass